ABSTRACT: AIMS/HYPOTHESIS:Type 2 diabetes is an established risk factor for heart failure, but age-specific data are sparse. We aimed to determine excess risk of heart failure, based on age, glycaemic control and kidney function in comparison with age- and sex-matched control individuals from the general population. METHODS:Individuals with type 2 diabetes registered in the Swedish National Diabetes Registry 1998-2012 (n?=?266,305) were compared with age-, sex- and county-matched control individuals without diabetes (n?=?1,323,504), and followed over a median of 5.6 years until 31 December 2013. RESULTS:We identified 266,305 individuals with type 2 diabetes (mean age 62.0 years, 45.3% women) and 1,323,504 control individuals. Of the individuals with type 2 diabetes and control individuals, 18,715 (7.0%) and 50,157 (3.8%) were hospitalised with a diagnosis of heart failure, respectively. Comparing individuals with diabetes with those in the control group, men and women with type 2 diabetes who were younger than 55 years of age had HRs for hospitalisation for heart failure of 2.07 (95% CI 1.73, 2.48) and 4.59 (95% CI 3.50, 6.02), respectively, using analyses adjusted for socioeconomic variables and associated conditions. Younger age, poorer glycaemic control and deteriorating renal function were all associated with increased excess risk of heart failure in those with type 2 diabetes compared with the control group. However, people with diabetes who were ?75 years and without albuminuria or with good glycaemic control (HbA1c ?52 mmol/mol [?6.9%]) had a similar risk of hospitalisation for heart failure as control individuals in the same age group. CONCLUSIONS/INTERPRETATION:Men and women aged <55 years with type 2 diabetes are at markedly elevated excess risk of heart failure. The excess risk declined with age, but persisted even with good glycaemic control. However, among those who were 75 years and older, diabetic individuals with well controlled glucose levels or without albuminuria had a risk of heart failure that was on a par with individuals without diabetes.
